Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Fusion-Data-Company/bristol-os --skill deep-researchgit clone --depth 1 https://github.com/Fusion-Data-Company/bristol-osWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/fusion-data-company/bristol-os/deep-research)<a href="https://agentmods.dev/skills/fusion-data-company/bristol-os/deep-research"><img src="https://agentmods.dev/badge/skills/fusion-data-company/bristol-os/deep-research/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/fusion-data-company/bristol-os/deep-research"><img src="https://agentmods.dev/badge/skills/fusion-data-company/bristol-os/deep-research.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00082 | $0.00641 |
| Opus 5 | $0.00041 | $0.00320 |
| Sonnet 5 | $0.00016 | $0.00128 |
| Haiku 4.5 | $0.00008 | $0.00064 |
Grade A, and why
deep-research scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 38 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Deep Research
Turn a plain-language question into a thorough, source-cited briefing. This is the research engine the other Bristol plays lean on.
When to use
The user wants more than a quick answer: "research the Nashville multifamily market," "what's happening with construction costs," "look into this developer," "find everything on this incentive program."
Tools, in order of preference
- Baked-in research keys (always available, no setup): load
bristol-os/keys.envand call Tavily, Exa, and Firecrawl directly — exact commands inbristol-os/docs/DATA-SOURCES.md. Also FRED/Census for macro & demographic facts, and Quarry (quarry-parcels) for property/owner facts. - Claude's built-in web search — always available.
- Yardi Matrix / finance connectors if connected (real estate / market data). Run the full method below regardless of which tools fire.
Method
- Clarify the decision. One sharp question if scope is unclear: what decision will this inform, and how deep do they need it? Then proceed.
- Decompose the question into 4–8 sub-questions. State them.
- Fan out: search each sub-question. Prefer primary and authoritative sources (government data, company filings, official market reports, reputable trade press). Note the date of every source.
- Triangulate: confirm important facts across 2+ independent sources. Flag anything you can only find once as "single-source — verify."
- Adversarially check: actively look for data that contradicts the emerging answer. Note disagreements rather than smoothing them over.
- Synthesize into the output format below. Lead with the answer.
- Cite everything with source + date. Never present an unsourced number as fact.
Output (save to the workspace, e.g. research/<topic>-<date>.md)
- Bottom line — the answer in 3–5 sentences.
- Key findings — the important points, each with a source and date.
- What's uncertain / contested — gaps, conflicting data, single-source items.
- So what for Bristol — implications for the deal/market/decision at hand.
- Sources — list with links and dates.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 38 lines · 82 tokens per session scan A d944c23690bd
deep-research is a skill published in the GitHub repository Fusion-Data-Company/bristol-os (1 stars, last pushed 2mo ago), licensed MIT. It adds 82 tokens to every session and 641 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
trade-execution
Trade execution using price action setups, major trend reversal trading top and bottom, MTR failure continuation, strong bull and bear breakout trading, strong and weak channel trading strategies, trading range strategies, opening range swings, integration with gold.technicalsetups for setup matching and execution…
synthesize
Thesis report synthesis — turn a thesis's markdown artifacts into its letter-size thesis-report.html, then OPTIMIZE it against real page renders. Four-step loop — pack (deterministic Python bundle of all sources + report/metrics.json), author (the LLM writes report/content.html — narrative, KPI tiles, badges…
trade-idea-generation
Systematic trade idea generation framework, fundamental quantitative and qualitative screening pipeline, catalyst-driven trade structuring, macro-to-micro idea translation, 20-60 day long-short portfolio management process.
chart-patterns
Chart pattern recognition, price action patterns, candlestick signal bars, pullback bar counting H1/H2/H3/H4, trend channels, micro channels, trading ranges, breakouts, major trend reversals 5-step sequence, wedges, double tops and bottoms, triangles, head and shoulders, climaxes, measured moves, support and…
challenge
Adversarial verification of research theses — cross-run/cross-thesis contradiction via the entity index, pre-mortem (Klarman/Kahneman: assume the loss already happened, reverse the path), inversion (Munger: invert, always invert), and wrongif falsifiability review. ≈50-finding cap sorted by severity; content-derived…
converge
Append-only gap closure and the cadence engine for research theses. Evaluates artifact CURRENT state (never git history, never task checkboxes) against current pins and wrongif falsifiers, then appends a Convergence section to tasks.md re-proposing the gaps. A clean run leaves tasks.md byte-identical. Deterministic…